In this application, we propose to elucidate the molecular mechanisms by which ATM suppresses oncogenic translocations in developing lymphocytes. Specifically, we will generate novel mouse models expressing mutated ATM protein (rather than loss ATM), identify and characterize genetic lesions that synergize with ATM mutation in lymphomagenesis, and identify agents that can preferentially target ATM-mutated cancer cells for therapy. Lymphoid malignancies are characterized by recurrent chromosome translocations that result from aberrant repair of DNA double-strand breaks that normally occur during lymphocyte development. The ATM kinase is a master regulator of the DNA damage responses and a tumor suppresser gene. Germline inactivation of ATM causes Ataxia-Telangiectasia (A-T) syndrome, a neurological disorder associated with primary immunodeficiency and greatly increased risk for lymphomas and leukemia. Somatic mutations of ATM have also been identified in a broad spectrum of human cancers including >50% of mantle cell lymphomas and nearly all T-cell prolymphocytic leukemia. Our preliminary study suggests that cancer-associated somatic ATM mutations are functionally distinct from those in A-T patients. While ~89% of A-T patients carry truncating mutations (frameshift, nonsense, splicing) that abrogate ATM protein expression, ~72% of cancer-associated ATM mutations in TCGA are missense mutations clustering around the kinase domain. Expression of catalytically-inactive ATM protein in hematopoietic stem cells is more oncogenic than loss of ATM, resulting in earlier and more frequent B and T cell lymphomas in mice. Based on these findings, we propose to 1) identify the mechanisms by which catalytically-inactive ATM protein promotes lymphomagenesis beyond loss of ATM; 2) evaluate the consequences of other recurrent ATM missense mutation found in human cancers; and 3) identify agents/targets that can preferentially target ATM-mutated cancers. The results from this study will further elucidate the functions of ATM in DNA repair and tumor suppressionand provide the rationale to selectively target ATM mutated human cancers. ATM missense mutations occur in 4-8% of common epithelial cancers (i.g. colon, bladders, pancreas etc.) in additional to lymphomas and are often associated with poor prognosis. The specific DNA repair defects and selective hypersensitivities identified in this study will provide the basis to target ATM-mutated cancer with selective chemotherapy inclinical trials and lead to new strategies to effectively manage cancer with defects in ATM and other DNA repair genes in the future. The similarity between ATM deficient murine thymic lymphomas and human T-ALL also provide the platform to identify and characterize prognostic markers for human T-ALL.